Imports¶

In [110]:
import pandas as pd
import seaborn as sns
import os
import matplotlib.pyplot as plt
import numpy as np
from sklearn.preprocessing import StandardScaler
from sklearn.preprocessing import LabelEncoder
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LinearRegression
from sklearn.metrics import mean_squared_error, r2_score
from sklearn.feature_selection import RFE
from sklearn.model_selection import GridSearchCV
import matplotlib.pyplot as plt
from sklearn.preprocessing import StandardScaler, LabelEncoder
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LogisticRegression
from sklearn.metrics import accuracy_score, precision_score, recall_score,balanced_accuracy_score
from sklearn.metrics import mean_squared_error,r2_score
from sklearn.metrics import confusion_matrix
import xgboost as xgb
from imblearn.under_sampling import RandomUnderSampler
from imblearn.over_sampling import SMOTE
from sklearn.metrics import classification_report
from imblearn.ensemble import BalancedRandomForestClassifier
import requests
from bs4 import BeautifulSoup
from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import RandomizedSearchCV
import requests
from bs4 import BeautifulSoup
import random
from ydata_profiling import ProfileReport
np.set_printoptions(threshold=np.inf)
In [111]:
# Définir le chemin du répertoire contenant vos fichiers CSV
directory = '/Users/stanislas/Downloads/To-Import-Model'

# Liste pour stocker les DataFrames de chaque fichier
dfs = []

# Parcourir tous les fichiers dans le répertoire
for filename in os.listdir(directory):
    if filename.endswith(".csv"):  # Assurez-vous que le fichier est un fichier CSV
        filepath = os.path.join(directory, filename)
        # Lire le fichier CSV dans un DataFrame et l'ajouter à la liste
        data = pd.read_csv(filepath)
        dfs.append(data)

# Concaténer tous les DataFrames dans un DataFrame global
df = pd.concat(dfs, ignore_index=True)
In [112]:
players = pd.read_csv('atp_players.csv')
In [113]:
top500=pd.read_excel('Top500-Modified.xlsx')

DF reprocessing¶

Basic information¶

In [90]:
df.dtypes
Out[90]:
tourney_id             object
tourney_name           object
surface                object
draw_size             float64
tourney_level          object
tourney_date            int64
match_num               int64
winner_id               int64
winner_seed           float64
winner_entry           object
winner_name            object
winner_hand            object
winner_ht             float64
winner_ioc             object
winner_age            float64
loser_id                int64
loser_seed            float64
loser_entry            object
loser_name             object
loser_hand             object
loser_ht              float64
loser_ioc              object
loser_age             float64
score                  object
best_of                 int64
round                  object
minutes               float64
w_ace                 float64
w_df                  float64
w_svpt                float64
w_1stIn               float64
w_1stWon              float64
w_2ndWon              float64
w_SvGms               float64
w_bpSaved             float64
w_bpFaced             float64
l_ace                 float64
l_df                  float64
l_svpt                float64
l_1stIn               float64
l_1stWon              float64
l_2ndWon              float64
l_SvGms               float64
l_bpSaved             float64
l_bpFaced             float64
winner_rank           float64
winner_rank_points    float64
loser_rank            float64
loser_rank_points     float64
dtype: object
In [91]:
df.describe()
Out[91]:
draw_size tourney_date match_num winner_id winner_seed winner_ht winner_age loser_id loser_seed loser_ht ... l_1stIn l_1stWon l_2ndWon l_SvGms l_bpSaved l_bpFaced winner_rank winner_rank_points loser_rank loser_rank_points
count 191085.000000 1.919200e+05 191920.000000 191920.000000 70926.000000 175012.000000 190609.000000 191920.000000 36099.000000 162257.000000 ... 95034.000000 95034.000000 95034.000000 95035.000000 95034.000000 95034.000000 156159.000000 108936.000000 147788.000000 107308.000000
mean 53.088479 1.993711e+07 79.398656 104499.014537 6.308279 184.491618 25.670153 105686.966012 7.711017 184.261696 ... 48.089421 31.962361 14.972936 12.206229 4.808942 8.730570 75.246960 1375.909663 112.905710 865.976908
std 36.645414 1.581156e+05 111.963129 13664.799449 5.547887 6.672384 4.053105 16655.971639 6.055301 6.658745 ... 19.378619 14.447415 7.207299 4.131171 3.275134 4.132557 120.956559 1729.720178 162.603579 993.837316
min 2.000000 1.967123e+07 1.000000 100001.000000 1.000000 160.000000 14.300000 100001.000000 1.000000 160.000000 ... 0.000000 0.000000 0.000000 0.000000 -6.000000 0.000000 1.000000 0.000000 1.000000 0.000000
25% 32.000000 1.980051e+07 11.000000 100417.000000 2.000000 180.000000 22.700000 100523.000000 4.000000 180.000000 ... 34.000000 22.000000 10.000000 9.000000 2.000000 6.000000 17.000000 494.000000 37.000000 365.000000
50% 32.000000 1.993052e+07 25.000000 101733.000000 5.000000 185.000000 25.300000 101900.000000 6.000000 185.000000 ... 45.000000 30.000000 14.000000 11.000000 4.000000 8.000000 45.000000 850.000000 70.000000 635.000000
75% 64.000000 2.007032e+07 94.000000 103990.000000 8.000000 188.000000 28.200000 104338.000000 10.000000 188.000000 ... 58.000000 40.000000 19.000000 15.000000 7.000000 11.000000 86.000000 1540.000000 118.000000 1015.000000
max 128.000000 2.023113e+07 1701.000000 212428.000000 35.000000 211.000000 58.700000 212437.000000 35.000000 211.000000 ... 328.000000 284.000000 101.000000 91.000000 28.000000 38.000000 2101.000000 16950.000000 2159.000000 16950.000000

8 rows × 35 columns

In [7]:
df.isna().sum()
Out[7]:
tourney_id                 0
tourney_name               0
surface                 2990
draw_size                835
tourney_level              0
tourney_date               0
match_num                  0
winner_id                  0
winner_seed           120994
winner_entry          175177
winner_name                0
winner_hand               12
winner_ht              16908
winner_ioc                 8
winner_age              1311
loser_id                   0
loser_seed            155821
loser_entry           163496
loser_name                 0
loser_hand                49
loser_ht               29663
loser_ioc                 72
loser_age               4657
score                      9
best_of                    0
round                      0
minutes                99653
w_ace                  96885
w_df                   96886
w_svpt                 96886
w_1stIn                96886
w_1stWon               96886
w_2ndWon               96886
w_SvGms                96885
w_bpSaved              96886
w_bpFaced              96886
l_ace                  96886
l_df                   96885
l_svpt                 96886
l_1stIn                96886
l_1stWon               96886
l_2ndWon               96886
l_SvGms                96885
l_bpSaved              96886
l_bpFaced              96886
winner_rank            35761
winner_rank_points     82984
loser_rank             44132
loser_rank_points      84612
dtype: int64

Keep only significant columns¶

In [10]:
df = df[['tourney_id','tourney_name','surface','draw_size','tourney_level','tourney_date',
         'match_num','winner_id','winner_hand','winner_ht','winner_age','loser_id','loser_hand',
        'loser_ht','loser_age','best_of','round']]

Further analysis¶

In [13]:
profile = ProfileReport(df, title="Rapport d'analyse exploratoire")
In [14]:
profile

Summarize dataset:   0%|          | 0/5 [00:00<?, ?it/s]
Generate report structure:   0%|          | 0/1 [00:00<?, ?it/s]
Render HTML:   0%|          | 0/1 [00:00<?, ?it/s]
Out[14]:




























Data Format¶






















Datetime format for tourney_date¶


















In [114]:



     


df['tourney_date']=df['tourney_date'].fillna(0)

# Convertir la colonne de flottants en chaînes de caractères
df['date_str'] = df['tourney_date'].astype(int).astype(str)

# Fonction pour convertir une chaîne de caractères en date, avec gestion des erreurs
def convert_to_date(date_str):
    try:
        return pd.to_datetime(date_str, format='%Y%m%d').strftime('%Y-%m-%d')
    except ValueError:
        return None  # Remplacer par NaN ou une valeur par défaut si nécessaire

# Appliquer la fonction à chaque valeur de la colonne et créer une nouvelle colonne 'date'
df['tourney_date'] = df['date_str'].apply(convert_to_date)

# Supprimer la colonne temporaire de chaînes de caractères
df.drop(columns=['date_str'], inplace=True)



     























Reducing values in the column tourney_name¶


















In [115]:



     


sorted(df['tourney_name'].unique())



     






















    
    
Out[115]:








['ATP Rio de Janeiro',
 'Aberavon',
 'Acapulco',
 'Adelaide',
 'Adelaide 1',
 'Adelaide 2',
 'Adelaide-2',
 'Aix en Provence',
 'Aix-en-Provence',
 'Alamo WCT',
 'Albany',
 'Algiers',
 'Amersfoort',
 'Amsterdam',
 'Amsterdam WCT',
 'Anaheim',
 'Ancona',
 'Antalya',
 'Antwerp',
 'Aptos',
 'Astana',
 'Athens',
 'Athens Olympics',
 'Atlanta',
 'Atlanta Olympics',
 'Atlanta WCT',
 'Atp Cup',
 'Auckland',
 'Australian Chps.',
 'Australian Open',
 'Australian Open-2',
 'Australian Round Robin',
 'Aviles',
 'Bahia',
 'Bakersfield WCT',
 'Baltimore',
 'Baltimore WCT',
 'Bangalore',
 'Bangkok',
 'Banja Luka',
 'Barcelona',
 'Barcelona 2',
 'Barcelona Nationals',
 'Barcelona Olympics',
 'Barcelona WCT',
 'Bari',
 'Barranquilla',
 'Basel',
 'Bastad',
 'Bastad 1',
 'Bastad WCT',
 'Beckenham',
 'Beijing',
 'Beijing Olympics',
 'Belgrade',
 'Belgrade ',
 'Belgrade 2',
 'Berkeley',
 'Berlin',
 'Bermuda',
 'Binghamton',
 'Binghamton NTL',
 'Birmingham',
 'Birmingham WCT',
 'Bloemfontein',
 'Boca Raton',
 'Boca West',
 'Bogota',
 'Bogota NTL',
 'Bologna',
 'Bologna WCT',
 'Bolzano',
 'Bombay',
 'Bordeaux',
 'Boston',
 'Boston 1',
 'Boston 2',
 'Boston WCT',
 'Bournemouth',
 'Brasilia',
 'Bremen Indoors',
 'Bretton Woods',
 'Brighton',
 'Brisbane',
 'Bristol',
 'Brussels',
 'Brussels WCT',
 'Bucharest',
 'Budapest',
 'Buenos Aires',
 'Buenos Aires NTL',
 'Buffalo',
 'Buffalo WCT',
 'Buzios',
 'Cagliari',
 'Cairo',
 'Calcutta',
 'Calgary',
 'Cambridge',
 'Canada Masters',
 'Cannes',
 'Cannes Chps',
 'Cannes WCT',
 "Cap D'Adge WCT",
 'Cape Town',
 'Cape Town WCT',
 'Caracas',
 'Caracas WCT',
 'Casablanca',
 'Casablanca WCT',
 'Catania',
 'Cedar Grove',
 'Champions Classic',
 'Charleston',
 'Charlotte',
 'Charlotte WCT',
 'Chengdu',
 'Chennai',
 'Chicago',
 'Chicago WCT',
 'Chicago-2 WCT',
 'Christchurch',
 'Cincinnati',
 'Cincinnati Masters',
 'Cleveland',
 'Cleveland WCT',
 'Cologne',
 'Cologne 1',
 'Cologne 2',
 'Cologne WCT',
 'Colombus',
 'Columbia',
 'Columbus',
 'Columbus WCT',
 'Concord Indoors',
 'Copenhagen',
 'Copenhagen WCT',
 'Coral Springs',
 'Cordoba',
 'Corpus Christi',
 'Corpus Christi NTL',
 'Costa Do Sauipe',
 'Curacao',
 'Dallas',
 'Dallas WCT',
 'Davis Cup AFR QF: ALG vs CIV',
 'Davis Cup AFR QF: CIV vs NGR',
 'Davis Cup AFR QF: KEN vs ZIM',
 'Davis Cup AFR QF: MAR vs KEN',
 'Davis Cup AFR QF: MEX vs KOR',
 'Davis Cup AFR QF: SEN vs ALG',
 'Davis Cup AFR QF: SEN vs NGR',
 'Davis Cup AFR QF: ZIM vs LBA',
 'Davis Cup AFR R1: CIV vs TUN',
 'Davis Cup AFR R1: SEN vs MAR',
 'Davis Cup AFR R1: TUN vs NGR',
 'Davis Cup AFR SF: ALG vs NGR',
 'Davis Cup AFR SF: NZL vs KOR',
 'Davis Cup AFR SF: SEN vs EGY',
 'Davis Cup AFR SF: ZIM vs MAR',
 'Davis Cup AFR SF: ZIM vs NGR',
 'Davis Cup AME F: ARG vs USA',
 'Davis Cup AME F: BRA vs CAN',
 'Davis Cup AME F: BRA vs ECU',
 'Davis Cup AME F: BRA vs MEX',
 'Davis Cup AME F: CAN vs COL',
 'Davis Cup AME F: CAN vs NZL',
 'Davis Cup AME F: CAN vs PAR',
 'Davis Cup AME F: CHI vs ARG',
 'Davis Cup AME F: CHI vs BRA',
 'Davis Cup AME F: CHI vs RSA',
 'Davis Cup AME F: COL vs USA',
 'Davis Cup AME F: COL vs VEN',
 'Davis Cup AME F: ECU vs BRA',
 'Davis Cup AME F: JPN vs AUS',
 'Davis Cup AME F: JPN vs PHI',
 'Davis Cup AME F: MEX vs AUS',
 'Davis Cup AME F: MEX vs BRA',
 'Davis Cup AME F: MEX vs NZL',
 'Davis Cup AME F: MEX vs USA',
 'Davis Cup AME F: USA vs ARG',
 'Davis Cup AME F: USA vs COL',
 'Davis Cup AME F: USA vs ECU',
 'Davis Cup AME F: USA vs MEX',
 'Davis Cup AME F: USA vs RSA',
 'Davis Cup AME PR: ARG vs ECU',
 'Davis Cup AME PR: BOL vs PER',
 'Davis Cup AME PR: BRA vs BOL',
 'Davis Cup AME PR: BRA vs PER',
 'Davis Cup AME PR: CAN vs CAR',
 'Davis Cup AME PR: CAN vs COL',
 'Davis Cup AME PR: CAN vs MEX',
 'Davis Cup AME PR: CAR vs CAN',
 'Davis Cup AME PR: CAR vs USA',
 'Davis Cup AME PR: CAR vs VEN',
 'Davis Cup AME PR: COL vs CAR',
 'Davis Cup AME PR: COL vs VEN',
 'Davis Cup AME PR: ECU vs ARG',
 'Davis Cup AME PR: MEX vs CAN',
 'Davis Cup AME PR: MEX vs CAR',
 'Davis Cup AME PR: PER vs URU',
 'Davis Cup AME PR: PHI vs JPN',
 'Davis Cup AME PR: URU vs ECU',
 'Davis Cup AME PR: USA vs MEX',
 'Davis Cup AME PR: USA vs VEN',
 'Davis Cup AME PR: VEN vs COL',
 'Davis Cup AME PR: VEN vs USA',
 'Davis Cup AME QF: ARG vs BRA',
 'Davis Cup AME QF: ARG vs URU',
 'Davis Cup AME QF: BRA vs ARG',
 'Davis Cup AME QF: BRA vs PER',
 'Davis Cup AME QF: CAN vs CAR',
 'Davis Cup AME QF: CAN vs VEN',
 'Davis Cup AME QF: CAR vs BRA',
 'Davis Cup AME QF: CAR vs CHI',
 'Davis Cup AME QF: CAR vs URU',
 'Davis Cup AME QF: CHI vs CAN',
 'Davis Cup AME QF: CHI vs COL',
 'Davis Cup AME QF: COL vs BRA',
 'Davis Cup AME QF: COL vs PER',
 'Davis Cup AME QF: ECU vs BOL',
 'Davis Cup AME QF: ECU vs CAR',
 'Davis Cup AME QF: HKG vs AUS',
 'Davis Cup AME QF: JPN vs CHN',
 'Davis Cup AME QF: JPN vs PHI',
 'Davis Cup AME QF: MEX vs CAN',
 'Davis Cup AME QF: MEX vs PER',
 'Davis Cup AME QF: MEX vs USA',
 'Davis Cup AME QF: PER vs CAN',
 'Davis Cup AME QF: URU vs BRA',
 'Davis Cup AME QF: URU vs COL',
 'Davis Cup AME QF: URU vs MEX',
 'Davis Cup AME QF: USA vs MEX',
 'Davis Cup AME QF: VEN vs BRA',
 'Davis Cup AME R1: CAR vs CUB',
 'Davis Cup AME R1: COL vs URU',
 'Davis Cup AME R1: PER vs BRA',
 'Davis Cup AME R1: VEN vs CAN',
 'Davis Cup AME R1: VEN vs MEX',
 'Davis Cup AME R1: VEN vs URU',
 'Davis Cup AME SF: ARG vs CHI',
 'Davis Cup AME SF: ARG vs PER',
 'Davis Cup AME SF: AUS vs INA',
 'Davis Cup AME SF: AUS vs JPN',
 'Davis Cup AME SF: BRA vs CHI',
 'Davis Cup AME SF: BRA vs COL',
 'Davis Cup AME SF: BRA vs URU',
 'Davis Cup AME SF: CAN vs CAR',
 'Davis Cup AME SF: CAN vs ECU',
 'Davis Cup AME SF: CAN vs MEX',
 'Davis Cup AME SF: CAR vs MEX',
 'Davis Cup AME SF: CAR vs NZL',
 'Davis Cup AME SF: CAR vs USA',
 'Davis Cup AME SF: CHI vs ARG',
 'Davis Cup AME SF: CHI vs BRA',
 'Davis Cup AME SF: CHI vs MEX',
 'Davis Cup AME SF: CHN vs JPN',
 'Davis Cup AME SF: COL vs CAN',
 'Davis Cup AME SF: COL vs CAR',
 'Davis Cup AME SF: ECU vs CAN',
 'Davis Cup AME SF: INA vs AUS',
 'Davis Cup AME SF: JPN vs AUS',
 'Davis Cup AME SF: JPN vs PHI',
 'Davis Cup AME SF: KOR vs AUS',
 'Davis Cup AME SF: MEX vs CAN',
 'Davis Cup AME SF: MEX vs NZL',
 'Davis Cup AME SF: PHI vs JPN',
 'Davis Cup AME SF: URU vs BRA',
 'Davis Cup AME SF: USA vs CAR',
 'Davis Cup AME SF: USA vs RSA',
 'Davis Cup AME SF: VEN vs CAN',
 'Davis Cup AME SF: VEN vs CAR',
 'Davis Cup AOC F: NZL vs KOR',
 'Davis Cup AOC QF: KOR vs PAK',
 'Davis Cup AOC QF: KOR vs PHI',
 'Davis Cup AOC QF: PAK vs INA',
 'Davis Cup AOC QF: PAK vs KOR',
 'Davis Cup AOC QF: PHI vs KOR',
 'Davis Cup AOC QF: SRI vs IND',
 'Davis Cup AOC R1: INA vs PHI',
 'Davis Cup AOC SF: INA vs PAK',
 'Davis Cup AOC SF: IND vs THA',
 'Davis Cup AOC SF: PHI vs KOR',
 'Davis Cup AOC SF: THA vs IND',
 'Davis Cup EUR F: AUS vs NZL',
 'Davis Cup EUR F: BRA vs ROU',
 'Davis Cup EUR F: CHN vs NZL',
 'Davis Cup EUR F: COL vs RSA',
 'Davis Cup EUR F: DEN vs AUT',
 'Davis Cup EUR F: ESP vs HUN',
 'Davis Cup EUR F: ESP vs ITA',
 'Davis Cup EUR F: ESP vs SWE',
 'Davis Cup EUR F: ESP vs TCH',
 'Davis Cup EUR F: ESP vs YUG',
 'Davis Cup EUR F: FRA vs AUT',
 'Davis Cup EUR F: FRA vs ROU',
 'Davis Cup EUR F: FRG vs ESP',
 'Davis Cup EUR F: FRG vs RSA',
 'Davis Cup EUR F: FRG vs SUI',
 'Davis Cup EUR F: FRG vs URS',
 'Davis Cup EUR F: GBR vs ISR',
 'Davis Cup EUR F: GBR vs ITA',
 'Davis Cup EUR F: GBR vs ROU',
 'Davis Cup EUR F: GBR vs RSA',
 'Davis Cup EUR F: GBR vs TCH',
 'Davis Cup EUR F: HUN vs DEN',
 'Davis Cup EUR F: HUN vs ESP',
 'Davis Cup EUR F: IND vs JPN',
 'Davis Cup EUR F: IRL vs SUI',
 'Davis Cup EUR F: ITA vs GBR',
 'Davis Cup EUR F: ITA vs ROU',
 'Davis Cup EUR F: ITA vs SWE',
 'Davis Cup EUR F: JPN vs IND',
 'Davis Cup EUR F: NZL vs AUS',
 'Davis Cup EUR F: PAK vs JPN',
 'Davis Cup EUR F: ROU vs DEN',
 'Davis Cup EUR F: ROU vs FRG',
 'Davis Cup EUR F: ROU vs TCH',
 'Davis Cup EUR F: ROU vs URS',
 'Davis Cup EUR F: SUI vs ISR',
 'Davis Cup EUR F: SWE vs HUN',
 'Davis Cup EUR F: TCH vs ESP',
 'Davis Cup EUR F: TCH vs FRA',
 'Davis Cup EUR F: TCH vs ITA',
 'Davis Cup EUR F: TCH vs SWE',
 'Davis Cup EUR F: URS vs HUN',
 'Davis Cup EUR F: URS vs ISR',
 'Davis Cup EUR F: URS vs NED',
 'Davis Cup EUR F: URS vs ROU',
 'Davis Cup EUR F: URS vs SUI',
 'Davis Cup EUR F: URS vs TCH',
 'Davis Cup EUR F: YUG vs HUN',
 'Davis Cup EUR PQ: ALG vs MAR',
 'Davis Cup EUR PQ: EGY vs MAR',
 'Davis Cup EUR PQ: IRI vs ALG',
 'Davis Cup EUR PQ: IRI vs ISR',
 'Davis Cup EUR PQ: IRI vs LIB',
 'Davis Cup EUR PQ: ISR vs LUX',
 'Davis Cup EUR PQ: ISR vs TUR',
 'Davis Cup EUR PQ: LUX vs FIN',
 'Davis Cup EUR PQ: LUX vs POR',
 'Davis Cup EUR PQ: LUX vs TUR',
 'Davis Cup EUR PQ: MAR vs TUR',
 'Davis Cup EUR PQ: NGR vs KEN',
 'Davis Cup EUR PQ: NGR vs MAR',
 'Davis Cup EUR PQ: PHI vs PAK',
 'Davis Cup EUR PQ: TUR vs IRI',
 'Davis Cup EUR PQ: TUR vs IRL',
 'Davis Cup EUR PQ: TUR vs LIB',
 'Davis Cup EUR PQ: TUR vs LUX',
 'Davis Cup EUR PR: ALG vs IRI',
 'Davis Cup EUR PR: AUT vs EGY',
 'Davis Cup EUR PR: AUT vs FIN',
 'Davis Cup EUR PR: AUT vs GBR',
 'Davis Cup EUR PR: AUT vs MON',
 'Davis Cup EUR PR: AUT vs NZL',
 'Davis Cup EUR PR: AUT vs ROU',
 'Davis Cup EUR PR: BEL vs AUT',
 'Davis Cup EUR PR: BEL vs BUL',
 'Davis Cup EUR PR: BEL vs DEN',
 'Davis Cup EUR PR: BEL vs GRE',
 'Davis Cup EUR PR: BEL vs HUN',
 'Davis Cup EUR PR: BEL vs IRL',
 'Davis Cup EUR PR: BEL vs NED',
 'Davis Cup EUR PR: BEL vs NOR',
 'Davis Cup EUR PR: BEL vs ROU',
 'Davis Cup EUR PR: BEL vs YUG',
 'Davis Cup EUR PR: BUL vs AUT',
 'Davis Cup EUR PR: BUL vs HUN',
 'Davis Cup EUR PR: BUL vs IRI',
 'Davis Cup EUR PR: BUL vs TUR',
 'Davis Cup EUR PR: DEN vs FIN',
 'Davis Cup EUR PR: DEN vs GRE',
 'Davis Cup EUR PR: EGY vs BUL',
 'Davis Cup EUR PR: EGY vs GBR',
 'Davis Cup EUR PR: EGY vs IRL',
 'Davis Cup EUR PR: EGY vs LIB',
 'Davis Cup EUR PR: EGY vs POR',
 'Davis Cup EUR PR: EGY vs TUR',
 'Davis Cup EUR PR: ESP vs DEN',
 'Davis Cup EUR PR: ESP vs NED',
 'Davis Cup EUR PR: ESP vs NOR',
 'Davis Cup EUR PR: ESP vs URS',
 'Davis Cup EUR PR: FIN vs EGY',
 'Davis Cup EUR PR: FIN vs MAR',
 'Davis Cup EUR PR: FIN vs POL',
 'Davis Cup EUR PR: FIN vs TUR',
 'Davis Cup EUR PR: FRA vs BEL',
 'Davis Cup EUR PR: FRA vs URS',
 'Davis Cup EUR PR: FRG vs DEN',
 'Davis Cup EUR PR: FRG vs ISR',
 'Davis Cup EUR PR: FRG vs NOR',
 'Davis Cup EUR PR: FRG vs SUI',
 'Davis Cup EUR PR: GBR vs IRI',
 'Davis Cup EUR PR: GRE vs DEN',
 'Davis Cup EUR PR: GRE vs ESP',
 'Davis Cup EUR PR: GRE vs FIN',
 'Davis Cup EUR PR: GRE vs POR',
 'Davis Cup EUR PR: GRE vs URS',
 'Davis Cup EUR PR: HUN vs BEL',
 'Davis Cup EUR PR: HUN vs NED',
 'Davis Cup EUR PR: IRI vs SUI',
 'Davis Cup EUR PR: IRL vs BUL',
 'Davis Cup EUR PR: IRL vs IRI',
 'Davis Cup EUR PR: IRL vs NOR',
 'Davis Cup EUR PR: IRL vs POR',
 'Davis Cup EUR PR: IRL vs SWE',
 'Davis Cup EUR PR: ISR vs AUT',
 'Davis Cup EUR PR: ISR vs FIN',
 'Davis Cup EUR PR: ISR vs IRI',
 'Davis Cup EUR PR: ISR vs MON',
 'Davis Cup EUR PR: ISR vs SUI',
 'Davis Cup EUR PR: ITA vs DEN',
 'Davis Cup EUR PR: ITA vs POL',
 'Davis Cup EUR PR: JPN vs IND',
 'Davis Cup EUR PR: MAR vs BEL',
 'Davis Cup EUR PR: MAR vs LUX',
 'Davis Cup EUR PR: MAR vs NOR',
 'Davis Cup EUR PR: MAR vs YUG',
 'Davis Cup EUR PR: MON vs DEN',
 'Davis Cup EUR PR: MON vs EGY',
 'Davis Cup EUR PR: MON vs GBR',
 'Davis Cup EUR PR: MON vs ISR',
 'Davis Cup EUR PR: MON vs LUX',
 'Davis Cup EUR PR: MON vs NGR',
 'Davis Cup EUR PR: MON vs URS',
 'Davis Cup EUR PR: NED vs DEN',
 'Davis Cup EUR PR: NED vs FIN',
 'Davis Cup EUR PR: NED vs FRA',
 'Davis Cup EUR PR: NED vs GRE',
 'Davis Cup EUR PR: NED vs ISR',
 'Davis Cup EUR PR: NED vs NOR',
 'Davis Cup EUR PR: NED vs YUG',
 'Davis Cup EUR PR: NOR vs IRI',
 'Davis Cup EUR PR: NOR vs TUR',
 'Davis Cup EUR PR: PHI vs THA',
 'Davis Cup EUR PR: POL vs FIN',
 'Davis Cup EUR PR: POL vs FRG',
 'Davis Cup EUR PR: POL vs HUN',
 'Davis Cup EUR PR: POL vs IRI',
 'Davis Cup EUR PR: POL vs NOR',
 'Davis Cup EUR PR: POL vs POR',
 'Davis Cup EUR PR: POR vs FRA',
 'Davis Cup EUR PR: POR vs IRL',
 'Davis Cup EUR PR: POR vs MON',
 'Davis Cup EUR PR: POR vs SUI',
 'Davis Cup EUR PR: ROU vs BEL',
 'Davis Cup EUR PR: SRI vs PAK',
 'Davis Cup EUR PR: SUI vs AUT',
 'Davis Cup EUR PR: SUI vs EGY',
 'Davis Cup EUR PR: SUI vs FRA',
 'Davis Cup EUR PR: SUI vs FRG',
 'Davis Cup EUR PR: SUI vs GBR',
 'Davis Cup EUR PR: SUI vs IRI',
 'Davis Cup EUR PR: SUI vs ISR',
 'Davis Cup EUR PR: SUI vs RHO',
 'Davis Cup EUR PR: SUI vs YUG',
 'Davis Cup EUR PR: SWE vs MON',
 'Davis Cup EUR PR: SWE vs POL',
 'Davis Cup EUR PR: TCH vs IRL',
 'Davis Cup EUR PR: TCH vs NED',
 'Davis Cup EUR PR: TPE vs PHI',
 'Davis Cup EUR PR: URS vs MON',
 'Davis Cup EUR PR: YUG vs BUL',
 'Davis Cup EUR PR: YUG vs GRE',
 'Davis Cup EUR PR: YUG vs POR',
 'Davis Cup EUR PR: YUG vs ROU',
 'Davis Cup EUR QF: ALG vs ESP',
 'Davis Cup EUR QF: AUT vs ALG',
 'Davis Cup EUR QF: AUT vs DEN',
 'Davis Cup EUR QF: AUT vs ESP',
 'Davis Cup EUR QF: AUT vs FRA',
 'Davis Cup EUR QF: AUT vs GRE',
 'Davis Cup EUR QF: AUT vs NOR',
 'Davis Cup EUR QF: AUT vs POR',
 'Davis Cup EUR QF: BEL vs FRG',
 'Davis Cup EUR QF: BEL vs ISR',
 'Davis Cup EUR QF: BEL vs TCH',
 'Davis Cup EUR QF: BEL vs URS',
 'Davis Cup EUR QF: BUL vs BEL',
 'Davis Cup EUR QF: BUL vs EGY',
 'Davis Cup EUR QF: BUL vs FIN',
 'Davis Cup EUR QF: BUL vs FRG',
 'Davis Cup EUR QF: BUL vs SEN',
 'Davis Cup EUR QF: DEN vs BEL',
 'Davis Cup EUR QF: DEN vs NED',
 'Davis Cup EUR QF: DEN vs POL',
 'Davis Cup EUR QF: EGY vs HUN',
 'Davis Cup EUR QF: EGY vs NED',
 'Davis Cup EUR QF: EGY vs TCH',
 'Davis Cup EUR QF: EGY vs YUG',
 'Davis Cup EUR QF: ESP vs BUL',
 'Davis Cup EUR QF: ESP vs FRG',
 'Davis Cup EUR QF: ESP vs GBR',
 'Davis Cup EUR QF: ESP vs SWE',
 'Davis Cup EUR QF: ESP vs YUG',
 'Davis Cup EUR QF: FIN vs BUL',
 'Davis Cup EUR QF: FIN vs DEN',
 'Davis Cup EUR QF: FIN vs IRL',
 'Davis Cup EUR QF: FRA vs AUT',
 'Davis Cup EUR QF: FRA vs ESP',
 'Davis Cup EUR QF: FRA vs FIN',
 'Davis Cup EUR QF: FRA vs SUI',
 'Davis Cup EUR QF: FRA vs YUG',
 'Davis Cup EUR QF: FRG vs AUT',
 'Davis Cup EUR QF: FRG vs EGY',
 'Davis Cup EUR QF: FRG vs ESP',
 'Davis Cup EUR QF: FRG vs GBR',
 'Davis Cup EUR QF: FRG vs IRL',
 'Davis Cup EUR QF: FRG vs SWE',
 'Davis Cup EUR QF: FRG vs URS',
 'Davis Cup EUR QF: GBR vs AUT',
 'Davis Cup EUR QF: GBR vs IRL',
 'Davis Cup EUR QF: GBR vs POR',
 'Davis Cup EUR QF: GBR vs ROU',
 'Davis Cup EUR QF: GRE vs AUT',
 'Davis Cup EUR QF: GRE vs FIN',
 'Davis Cup EUR QF: HUN vs EGY',
 'Davis Cup EUR QF: HUN vs FRG',
 'Davis Cup EUR QF: HUN vs LUX',
 'Davis Cup EUR QF: HUN vs URS',
 'Davis Cup EUR QF: HUN vs YUG',
 'Davis Cup EUR QF: HUN vs ZIM',
 'Davis Cup EUR QF: IND vs JPN',
 'Davis Cup EUR QF: IRI vs RSA',
 'Davis Cup EUR QF: IRL vs MON',
 'Davis Cup EUR QF: IRL vs NED',
 'Davis Cup EUR QF: IRL vs YUG',
 'Davis Cup EUR QF: ISR vs BEL',
 'Davis Cup EUR QF: ISR vs NED',
 'Davis Cup EUR QF: ISR vs POL',
 'Davis Cup EUR QF: ISR vs ROU',
 'Davis Cup EUR QF: ITA vs AUT',
 'Davis Cup EUR QF: ITA vs BUL',
 'Davis Cup EUR QF: ITA vs MON',
 'Davis Cup EUR QF: ITA vs NED',
 'Davis Cup EUR QF: ITA vs YUG',
 'Davis Cup EUR QF: JPN vs PAK',
 'Davis Cup EUR QF: MAR vs SUI',
 'Davis Cup EUR QF: MAR vs URS',
 'Davis Cup EUR QF: MON vs HUN',
 'Davis Cup EUR QF: MON vs IRL',
 'Davis Cup EUR QF: MON vs ISR',
 'Davis Cup EUR QF: MON vs POL',
 'Davis Cup EUR QF: MON vs POR',
 'Davis Cup EUR QF: MON vs TCH',
 'Davis Cup EUR QF: MON vs URS',
 'Davis Cup EUR QF: NED vs ESP',
 'Davis Cup EUR QF: NED vs IRL',
 'Davis Cup EUR QF: NED vs ROU',
 'Davis Cup EUR QF: NGR vs NED',
 'Davis Cup EUR QF: NOR vs AUT',
 'Davis Cup EUR QF: NOR vs FRA',
 'Davis Cup EUR QF: NOR vs ROU',
 'Davis Cup EUR QF: PAK vs PHI',
 'Davis Cup EUR QF: PHI vs IND',
 'Davis Cup EUR QF: POL vs FRA',
 'Davis Cup EUR QF: POL vs ITA',
 'Davis Cup EUR QF: POL vs ROU',
 'Davis Cup EUR QF: POR vs EGY',
 'Davis Cup EUR QF: POR vs HUN',
 'Davis Cup EUR QF: ROU vs AUT',
 'Davis Cup EUR QF: ROU vs FRG',
 'Davis Cup EUR QF: ROU vs GRE',
 'Davis Cup EUR QF: ROU vs IRI',
 'Davis Cup EUR QF: ROU vs ISR',
 'Davis Cup EUR QF: ROU vs POL',
 'Davis Cup EUR QF: ROU vs TCH',
 'Davis Cup EUR QF: SUI vs BEL',
 'Davis Cup EUR QF: SUI vs ESP',
 'Davis Cup EUR QF: SUI vs GRE',
 'Davis Cup EUR QF: SUI vs HUN',
 'Davis Cup EUR QF: SUI vs SEN',
 'Davis Cup EUR QF: SUI vs ZIM',
 'Davis Cup EUR QF: SWE vs AUT',
 'Davis Cup EUR QF: SWE vs ESP',
 'Davis Cup EUR QF: SWE vs FRG',
 'Davis Cup EUR QF: SWE vs NED',
 'Davis Cup EUR QF: TCH vs POL',
 'Davis Cup EUR QF: TCH vs POR',
 'Davis Cup EUR QF: TCH vs SWE',
 'Davis Cup EUR QF: THA vs PAK',
 'Davis Cup EUR QF: TPE vs NZL',
 'Davis Cup EUR QF: TUR vs FRA',
 'Davis Cup EUR QF: TUR vs ROU',
 'Davis Cup EUR QF: TUR vs URS',
 'Davis Cup EUR QF: URS vs BEL',
 'Davis Cup EUR QF: URS vs CAN',
 'Davis Cup EUR QF: URS vs MON',
 'Davis Cup EUR QF: URS vs YUG',
 'Davis Cup EUR QF: YUG vs ESP',
 'Davis Cup EUR QF: YUG vs ISR',
 'Davis Cup EUR QF: YUG vs ITA',
 'Davis Cup EUR QF: YUG vs NZL',
 'Davis Cup EUR QF: YUG vs SWE',
 'Davis Cup EUR R1: ALG vs BUL',
 'Davis Cup EUR R1: AUT vs MAR',
 'Davis Cup EUR R1: AUT vs NZL',
 'Davis Cup EUR R1: AUT vs RSA',
 'Davis Cup EUR R1: BEL vs BUL',
 'Davis Cup EUR R1: BEL vs POL',
 'Davis Cup EUR R1: BEL vs TCH',
 'Davis Cup EUR R1: BUL vs BEL',
 'Davis Cup EUR R1: BUL vs CYP',
 'Davis Cup EUR R1: BUL vs DEN',
 'Davis Cup EUR R1: BUL vs ESP',
 'Davis Cup EUR R1: BUL vs TUR',
 'Davis Cup EUR R1: CHN vs INA',
 'Davis Cup EUR R1: CYP vs IRL',
 'Davis Cup EUR R1: DEN vs POR',
 'Davis Cup EUR R1: DEN vs TCH',
 'Davis Cup EUR R1: DEN vs URS',
 'Davis Cup EUR R1: EGY vs ALG',
 'Davis Cup EUR R1: EGY vs GRE',
 'Davis Cup EUR R1: EGY vs LUX',
 'Davis Cup EUR R1: EGY vs MLT',
 'Davis Cup EUR R1: EGY vs NOR',
 'Davis Cup EUR R1: EGY vs POL',
 'Davis Cup EUR R1: EGY vs TCH',
 'Davis Cup EUR R1: ESP vs ALG',
 'Davis Cup EUR R1: ESP vs NED',
 'Davis Cup EUR R1: ESP vs RHO',
 'Davis Cup EUR R1: FIN vs BEL',
 'Davis Cup EUR R1: FIN vs CYP',
 'Davis Cup EUR R1: FIN vs DEN',
 'Davis Cup EUR R1: FIN vs IRL',
 'Davis Cup EUR R1: FIN vs POR',
 'Davis Cup EUR R1: FIN vs SWE',
 'Davis Cup EUR R1: FRA vs GBR',
 'Davis Cup EUR R1: FRG vs DEN',
 'Davis Cup EUR R1: FRG vs NZL',
 'Davis Cup EUR R1: FRG vs SUI',
 'Davis Cup EUR R1: GBR vs AUT',
 'Davis Cup EUR R1: GBR vs FRA',
 'Davis Cup EUR R1: GRE vs BEL',
 'Davis Cup EUR R1: GRE vs FRG',
 'Davis Cup EUR R1: GRE vs HUN',
 'Davis Cup EUR R1: GRE vs LUX',
 'Davis Cup EUR R1: GRE vs NED',
 'Davis Cup EUR R1: GRE vs NOR',
 'Davis Cup EUR R1: GRE vs SYR',
 'Davis Cup EUR R1: GRE vs URS',
 'Davis Cup EUR R1: HUN vs POL',
 'Davis Cup EUR R1: HUN vs URS',
 'Davis Cup EUR R1: IRI vs EGY',
 'Davis Cup EUR R1: IRI vs ISR',
 'Davis Cup EUR R1: IRI vs ROU',
 'Davis Cup EUR R1: IRL vs BEL',
 'Davis Cup EUR R1: IRL vs LUX',
 'Davis Cup EUR R1: IRL vs TUR',
 'Davis Cup EUR R1: ISR vs NED',
 'Davis Cup EUR R1: ISR vs NOR',
 'Davis Cup EUR R1: ITA vs AUT',
 'Davis Cup EUR R1: ITA vs BEL',
 'Davis Cup EUR R1: ITA vs BUL',
 'Davis Cup EUR R1: ITA vs HUN',
 'Davis Cup EUR R1: ITA vs TCH',
 'Davis Cup EUR R1: LIB vs MAR',
 'Davis Cup EUR R1: LUX vs BUL',
 'Davis Cup EUR R1: LUX vs IRL',
 'Davis Cup EUR R1: LUX vs MON',
 'Davis Cup EUR R1: LUX vs NOR',
 'Davis Cup EUR R1: MAR vs FIN',
 'Davis Cup EUR R1: MAR vs HUN',
 'Davis Cup EUR R1: MAR vs POL',
 'Davis Cup EUR R1: MAR vs SWE',
 'Davis Cup EUR R1: MLT vs IRL',
 'Davis Cup EUR R1: MON vs BUL',
 'Davis Cup EUR R1: MON vs IRL',
 'Davis Cup EUR R1: MON vs LUX',
 'Davis Cup EUR R1: MON vs MAR',
 'Davis Cup EUR R1: MON vs POR',
 'Davis Cup EUR R1: MON vs ZIM',
 'Davis Cup EUR R1: NED vs CAN',
 'Davis Cup EUR R1: NED vs FIN',
 'Davis Cup EUR R1: NED vs NOR',
 'Davis Cup EUR R1: NGR vs NOR',
 'Davis Cup EUR R1: NOR vs AUT',
 'Davis Cup EUR R1: NOR vs DEN',
 'Davis Cup EUR R1: NOR vs IRL',
 'Davis Cup EUR R1: NOR vs POR',
 'Davis Cup EUR R1: NOR vs SEN',
 'Davis Cup EUR R1: PAK vs SRI',
 'Davis Cup EUR R1: POL vs BEL',
 'Davis Cup EUR R1: POL vs FIN',
 'Davis Cup EUR R1: POL vs GRE',
 'Davis Cup EUR R1: POL vs HUN',
 'Davis Cup EUR R1: POL vs YUG',
 'Davis Cup EUR R1: POL vs ZIM',
 'Davis Cup EUR R1: POR vs ISR',
 'Davis Cup EUR R1: POR vs LUX',
 'Davis Cup EUR R1: POR vs MON',
 'Davis Cup EUR R1: POR vs NED',
 'Davis Cup EUR R1: POR vs TUN',
 'Davis Cup EUR R1: POR vs TUR',
 'Davis Cup EUR R1: POR vs ZIM',
 'Davis Cup EUR R1: ROU vs DEN',
 'Davis Cup EUR R1: ROU vs EGY',
 'Davis Cup EUR R1: ROU vs NED',
 'Davis Cup EUR R1: ROU vs SUI',
 'Davis Cup EUR R1: RSA vs IRI',
 'Davis Cup EUR R1: SEN vs MON',
 'Davis Cup EUR R1: SEN vs TUN',
 'Davis Cup EUR R1: SUI vs FRA',
 'Davis Cup EUR R1: SUI vs FRG',
 'Davis Cup EUR R1: SUI vs GBR',
 'Davis Cup EUR R1: SWE vs ESP',
 'Davis Cup EUR R1: SWE vs FRA',
 'Davis Cup EUR R1: SWE vs NZL',
 'Davis Cup EUR R1: SWE vs RHO',
 'Davis Cup EUR R1: SYR vs TUR',
 'Davis Cup EUR R1: TCH vs BRA',
 'Davis Cup EUR R1: TUN vs SUI',
 'Davis Cup EUR R1: TUR vs BEL',
 'Davis Cup EUR R1: TUR vs BUL',
 'Davis Cup EUR R1: TUR vs GRE',
 'Davis Cup EUR R1: TUR vs LUX',
 'Davis Cup EUR R1: TUR vs ZIM',
 'Davis Cup EUR R1: URS vs GRE',
 'Davis Cup EUR R1: URS vs HUN',
 'Davis Cup EUR R1: YUG vs FRA',
 'Davis Cup EUR R1: YUG vs GBR',
 'Davis Cup EUR R1: YUG vs NOR',
 'Davis Cup EUR R1: YUG vs NZL',
 'Davis Cup EUR R1: YUG vs POL',
 'Davis Cup EUR R1: YUG vs TUN',
 'Davis Cup EUR R1: ZIM vs POL',
 'Davis Cup EUR SF: ARG vs AUS',
 'Davis Cup EUR SF: AUS vs TCH',
 'Davis Cup EUR SF: AUT vs HUN',
 'Davis Cup EUR SF: AUT vs ISR',
 'Davis Cup EUR SF: AUT vs ROU',
 'Davis Cup EUR SF: AUT vs SUI',
 'Davis Cup EUR SF: AUT vs URS',
 'Davis Cup EUR SF: BEL vs HUN',
 'Davis Cup EUR SF: BRA vs ESP',
 'Davis Cup EUR SF: BRA vs TCH',
 'Davis Cup EUR SF: BUL vs SUI',
 'Davis Cup EUR SF: BUL vs YUG',
 'Davis Cup EUR SF: COL vs MEX',
 'Davis Cup EUR SF: DEN vs EGY',
 'Davis Cup EUR SF: DEN vs ROU',
 'Davis Cup EUR SF: ESP vs FRA',
 'Davis Cup EUR SF: ESP vs GBR',
 'Davis Cup EUR SF: ESP vs IRL',
 'Davis Cup EUR SF: ESP vs MON',
 'Davis Cup EUR SF: ESP vs ROU',
 'Davis Cup EUR SF: ESP vs USA',
 'Davis Cup EUR SF: FIN vs NED',
 'Davis Cup EUR SF: FRA vs BUL',
 'Davis Cup EUR SF: FRA vs ESP',
 'Davis Cup EUR SF: FRA vs GBR',
 'Davis Cup EUR SF: FRA vs ITA',
 'Davis Cup EUR SF: FRA vs TCH',
 'Davis Cup EUR SF: FRG vs BEL',
 'Davis Cup EUR SF: FRG vs HUN',
 'Davis Cup EUR SF: FRG vs IND',
 'Davis Cup EUR SF: FRG vs TCH',
 'Davis Cup EUR SF: GBR vs BRA',
 'Davis Cup EUR SF: GBR vs ESP',
 'Davis Cup EUR SF: GBR vs FRA',
 'Davis Cup EUR SF: GBR vs FRG',
 'Davis Cup EUR SF: GBR vs ROU',
 'Davis Cup EUR SF: GBR vs SUI',
 'Davis Cup EUR SF: HUN vs ESP',
 'Davis Cup EUR SF: HUN vs ISR',
 'Davis Cup EUR SF: HUN vs ITA',
 'Davis Cup EUR SF: HUN vs SUI',
 'Davis Cup EUR SF: HUN vs TCH',
 'Davis Cup EUR SF: IND vs FRG',
 'Davis Cup EUR SF: IND vs JPN',
 'Davis Cup EUR SF: IND vs NZL',
 'Davis Cup EUR SF: IND vs PAK',
 'Davis Cup EUR SF: IRL vs FIN',
 'Davis Cup EUR SF: ISR vs FRG',
 'Davis Cup EUR SF: ISR vs HUN',
 'Davis Cup EUR SF: ISR vs SUI',
 'Davis Cup EUR SF: ITA vs ESP',
 'Davis Cup EUR SF: ITA vs FRA',
 'Davis Cup EUR SF: ITA vs HUN',
 'Davis Cup EUR SF: ITA vs SUI',
 'Davis Cup EUR SF: ITA vs SWE',
 'Davis Cup EUR SF: ITA vs TCH',
 'Davis Cup EUR SF: ITA vs URS',
 'Davis Cup EUR SF: JPN vs THA',
 'Davis Cup EUR SF: MEX vs COL',
 'Davis Cup EUR SF: MON vs DEN',
 'Davis Cup EUR SF: NED vs ISR',
 'Davis Cup EUR SF: NED vs URS',
 'Davis Cup EUR SF: NZL vs IND',
 'Davis Cup EUR SF: NZL vs JPN',
 'Davis Cup EUR SF: PAK vs THA',
 'Davis Cup EUR SF: POL vs URS',
 'Davis Cup EUR SF: POR vs AUT',
 'Davis Cup EUR SF: ROU vs EGY',
 'Davis Cup EUR SF: ROU vs FRA',
 'Davis Cup EUR SF: ROU vs GBR',
 'Davis Cup EUR SF: ROU vs ITA',
 'Davis Cup EUR SF: ROU vs NZL',
 'Davis Cup EUR SF: ROU vs SWE',
 'Davis Cup EUR SF: ROU vs YUG',
 'Davis Cup EUR SF: RSA vs COL',
 'Davis Cup EUR SF: RSA vs ITA',
 'Davis Cup EUR SF: SUI vs NED',
 'Davis Cup EUR SF: SWE vs CHI',
 'Davis Cup EUR SF: SWE vs ESP',
 'Davis Cup EUR SF: SWE vs FRG',
 'Davis Cup EUR SF: SWE vs ITA',
 'Davis Cup EUR SF: TCH vs AUS',
 'Davis Cup EUR SF: TCH vs FRA',
 'Davis Cup EUR SF: TCH vs FRG',
 'Davis Cup EUR SF: TCH vs HUN',
 'Davis Cup EUR SF: TCH vs ROU',
 'Davis Cup EUR SF: TCH vs URS',
 'Davis Cup EUR SF: THA vs NZL',
 'Davis Cup EUR SF: URS vs AUT',
 'Davis Cup EUR SF: URS vs ESP',
 'Davis Cup EUR SF: URS vs FRA',
 'Davis Cup EUR SF: URS vs ITA',
 'Davis Cup EUR SF: URS vs SWE',
 'Davis Cup EUR SF: URS vs TCH',
 'Davis Cup EUR SF: URS vs YUG',
 'Davis Cup EUR SF: USA vs ESP',
 'Davis Cup EUR SF: YUG vs ROU',
 'Davis Cup Eastern QF: CHN vs HKG',
 'Davis Cup Eastern QF: CHN vs KOR',
 'Davis Cup Eastern QF: HKG vs CHN',
 'Davis Cup Eastern QF: JPN vs PHI',
 'Davis Cup Eastern QF: NZL vs TPE',
 'Davis Cup Eastern QF: THA vs PHI',
 'Davis Cup Eastern SF: JPN vs CHN',
 'Davis Cup Eastern SF: NZL vs CHN',
 'Davis Cup Finals F: AUS vs ITA',
 'Davis Cup Finals F: CAN vs AUS',
 'Davis Cup Finals F: CAN vs ESP',
 'Davis Cup Finals F: RTF vs CRO',
 'Davis Cup Finals QF: ARG vs ESP',
 'Davis Cup Finals QF: AUS vs CAN',
 'Davis Cup Finals QF: AUS vs NED',
 'Davis Cup Finals QF: CAN vs FIN',
 'Davis Cup Finals QF: CRO vs ESP',
 'Davis Cup Finals QF: CZE vs AUS',
 'Davis Cup Finals QF: GBR vs GER',
 'Davis Cup Finals QF: GER vs CAN',
 'Davis Cup Finals QF: ITA vs CRO',
 'Davis Cup Finals QF: ITA vs NED',
 'Davis Cup Finals QF: ITA vs USA',
 'Davis Cup Finals QF: RTF vs SWE',
 'Davis Cup Finals QF: SRB vs GBR',
 'Davis Cup Finals QF: SRB vs KAZ',
 'Davis Cup Finals QF: SRB vs RUS',
 'Davis Cup Finals RR: ARG vs CHI',
 'Davis Cup Finals RR: ARG vs CRO',
 'Davis Cup Finals RR: ARG vs GER',
 'Davis Cup Finals RR: ARG vs SWE',
 'Davis Cup Finals RR: AUS vs BEL',
 'Davis Cup Finals RR: AUS vs COL',
 'Davis Cup Finals RR: AUS vs FRA',
 'Davis Cup Finals RR: AUS vs GBR',
 'Davis Cup Finals RR: AUS vs HUN',
 'Davis Cup Finals RR: AUS vs SUI',
 'Davis Cup Finals RR: BEL vs AUS',
 'Davis Cup Finals RR: BEL vs COL',
 'Davis Cup Finals RR: CAN vs CHI',
 'Davis Cup Finals RR: CAN vs ITA',
 'Davis Cup Finals RR: CAN vs KAZ',
 'Davis Cup Finals RR: CAN vs SWE',
 'Davis Cup Finals RR: CRO vs AUS',
 'Davis Cup Finals RR: CRO vs ESP',
 'Davis Cup Finals RR: CRO vs FIN',
 'Davis Cup Finals RR: CRO vs HUN',
 'Davis Cup Finals RR: CRO vs NED',
 'Davis Cup Finals RR: CRO vs RUS',
 'Davis Cup Finals RR: CRO vs SWE',
 'Davis Cup Finals RR: CRO vs USA',
 'Davis Cup Finals RR: CZE vs KOR',
 'Davis Cup Finals RR: ESP vs CAN',
 'Davis Cup Finals RR: ESP vs CZE',
 'Davis Cup Finals RR: ESP vs ECU',
 'Davis Cup Finals RR: ESP vs KOR',
 'Davis Cup Finals RR: ESP vs RTF',
 'Davis Cup Finals RR: ESP vs RUS',
 'Davis Cup Finals RR: ESP vs SRB',
 'Davis Cup Finals RR: FRA vs AUS',
 'Davis Cup Finals RR: FRA vs BEL',
 'Davis Cup Finals RR: FRA vs CZE',
 'Davis Cup Finals RR: FRA vs GBR',
 'Davis Cup Finals RR: FRA vs GER',
 'Davis Cup Finals RR: FRA vs JPN',
 'Davis Cup Finals RR: FRA vs SRB',
 'Davis Cup Finals RR: FRA vs SUI',
 'Davis Cup Finals RR: GBR vs CZE',
 'Davis Cup Finals RR: GBR vs FRA',
 'Davis Cup Finals RR: GBR vs KAZ',
 'Davis Cup Finals RR: GBR vs NED',
 'Davis Cup Finals RR: GBR vs SUI',
 'Davis Cup Finals RR: GER vs AUS',
 'Davis Cup Finals RR: GER vs AUT',
 'Davis Cup Finals RR: GER vs BEL',
 'Davis Cup Finals RR: GER vs CHI',
 'Davis Cup Finals RR: ITA vs ARG',
 'Davis Cup Finals RR: ITA vs CAN',
 'Davis Cup Finals RR: ITA vs CHI',
 'Davis Cup Finals RR: ITA vs COL',
 'Davis Cup Finals RR: ITA vs CRO',
 'Davis Cup Finals RR: ITA vs SWE',
 'Davis Cup Finals RR: KAZ vs NED',
 'Davis Cup Finals RR: KAZ vs SWE',
 'Davis Cup Finals RR: KOR vs CAN',
 'Davis Cup Finals RR: KOR vs SRB',
 'Davis Cup Finals RR: NED vs FIN',
 'Davis Cup Finals RR: NED vs USA',
 'Davis Cup Finals RR: RTF vs ECU',
 'Davis Cup Finals RR: SRB vs AUT',
 'Davis Cup Finals RR: SRB vs CAN',
 'Davis Cup Finals RR: SRB vs CZE',
 'Davis Cup Finals RR: SRB vs GER',
 'Davis Cup Finals RR: SRB vs JPN',
 'Davis Cup Finals RR: SRB vs KOR',
 'Davis Cup Finals RR: SWE vs CHI',
 'Davis Cup Finals RR: USA vs CAN',
 'Davis Cup Finals RR: USA vs COL',
 'Davis Cup Finals RR: USA vs FIN',
 'Davis Cup Finals RR: USA vs GBR',
 'Davis Cup Finals RR: USA vs ITA',
 'Davis Cup Finals RR: USA vs KAZ',
 'Davis Cup Finals RR: USA vs NED',
 'Davis Cup Finals SF: AUS vs CRO',
 'Davis Cup Finals SF: CRO vs SRB',
 'Davis Cup Finals SF: FIN vs AUS',
 'Davis Cup Finals SF: GBR vs ESP',
 'Davis Cup Finals SF: ITA vs CAN',
 'Davis Cup Finals SF: ITA vs SRB',
 'Davis Cup Finals SF: RTF vs GER',
 'Davis Cup Finals SF: RUS vs CAN',
 'Davis Cup G1 F: ARG vs USA',
 'Davis Cup G1 F: AUT vs GBR',
 'Davis Cup G1 F: CHI vs USA',
 'Davis Cup G1 F: INA vs IND',
 'Davis Cup G1 F: INA vs JPN',
 'Davis Cup G1 F: INA vs KOR',
 'Davis Cup G1 F: MAS vs IND',
 'Davis Cup G1 F: URS vs NED',
 'Davis Cup G1 F: USA vs CHI',
 'Davis Cup G1 PO: ARG vs ECU',
 'Davis Cup G1 PO: ARG vs VEN',
 'Davis Cup G1 PO: AUT vs DEN',
 'Davis Cup G1 PO: AUT vs UKR',
 'Davis Cup G1 PO: BAH vs CAN',
 'Davis Cup G1 PO: BAH vs COL',
 'Davis Cup G1 PO: BAH vs MEX',
 'Davis Cup G1 PO: BAH vs VEN',
 'Davis Cup G1 PO: BAR vs ECU',
 'Davis Cup G1 PO: BAR vs URU',
 'Davis Cup G1 PO: BLR vs GEO',
 'Davis Cup G1 PO: BLR vs MKD',
 'Davis Cup G1 PO: BLR vs NED',
 'Davis Cup G1 PO: BLR vs POL',
 'Davis Cup G1 PO: BLR vs POR',
 'Davis Cup G1 PO: BLR vs SVK',
 'Davis Cup G1 PO: BRA vs CHI',
 'Davis Cup G1 PO: BRA vs PER',
 'Davis Cup G1 PO: BUL vs ROU',
 'Davis Cup G1 PO: CAN vs ARG',
 'Davis Cup G1 PO: CAN vs CHI',
 'Davis Cup G1 PO: CAN vs DOM',
 'Davis Cup G1 PO: CAN vs VEN',
 'Davis Cup G1 PO: CHI vs CAN',
 'Davis Cup G1 PO: CHI vs COL',
 'Davis Cup G1 PO: CHI vs PER',
 'Davis Cup G1 PO: CHI vs URU',
 'Davis Cup G1 PO: CHN vs INA',
 'Davis Cup G1 PO: CHN vs PAK',
 'Davis Cup G1 PO: CHN vs THA',
 'Davis Cup G1 PO: CHN vs TPE',
 'Davis Cup G1 PO: COL vs ARG',
 'Davis Cup G1 PO: COL vs MEX',
 'Davis Cup G1 PO: CRO vs POR',
 'Davis Cup G1 PO: DEN vs ESP',
 'Davis Cup G1 PO: DOM vs CHI',
 'Davis Cup G1 PO: ECU vs BAH',
 'Davis Cup G1 PO: ECU vs MEX',
 'Davis Cup G1 PO: ECU vs PER',
 'Davis Cup G1 PO: FIN vs HUN',
 'Davis Cup G1 PO: FIN vs NOR',
 'Davis Cup G1 PO: FIN vs POL',
 'Davis Cup G1 PO: GBR vs ISR',
 'Davis Cup G1 PO: GBR vs POL',
 'Davis Cup G1 PO: GBR vs ROU',
 'Davis Cup G1 PO: GHA vs IRL',
 'Davis Cup G1 PO: HKG vs TPE',
 'Davis Cup G1 PO: HUN vs BEL',
 'Davis Cup G1 PO: HUN vs MAR',
 'Davis Cup G1 PO: HUN vs UKR',
 'Davis Cup G1 PO: INA vs CHN',
 'Davis Cup G1 PO: INA vs KOR',
 'Davis Cup G1 PO: INA vs LIB',
 'Davis Cup G1 PO: INA vs NZL',
 'Davis Cup G1 PO: INA vs TPE',
 'Davis Cup G1 PO: IND vs CHN',
 ...]
























In [116]:



     


def replace_tourney_name(df):
    mask1 = df['tourney_name'].str.contains('Davis Cup')
    df.loc[mask1, 'tourney_name'] = 'Davis Cup'
    
    mask2 = df['tourney_name'].str.contains('Olympics')
    df.loc[mask2, 'tourney_name'] = 'Olympics'
    
    mask3 = df['tourney_name'].str.contains('Australian Open')
    df.loc[mask3, 'tourney_name'] = 'Australian Open'
    
    mask4 = df['tourney_name'].str.contains('Adelaide')
    df.loc[mask4, 'tourney_name'] = 'Adelaide'
    
    mask5 = df['tourney_name'].str.contains('Washington')
    df.loc[mask5, 'tourney_name'] = 'Washington'
    
    mask6 = df['tourney_name'].str.contains('Barcelona')
    df.loc[mask6, 'tourney_name'] = 'Barcelona'
    
    mask7 = df['tourney_name'].str.contains('Belgrade')
    df.loc[mask7, 'tourney_name'] = 'Belgrade'
    
    mask8 = df['tourney_name'].str.contains('Cincinnati')
    df.loc[mask8, 'tourney_name'] = 'Cincinnati'
    
    mask9 = df['tourney_name'].str.contains('Hamburg')
    df.loc[mask9, 'tourney_name'] = 'Hamburg'
    
    mask10 = df['tourney_name'].str.contains('Indian Wells')
    df.loc[mask10, 'tourney_name'] = 'Indian Wells'
    
    mask11 = df['tourney_name'].str.contains('Madrid')
    df.loc[mask11, 'tourney_name'] = 'Madrid'
    
    mask12 = df['tourney_name'].str.contains('Miami')
    df.loc[mask12, 'tourney_name'] = 'Miami'
    
    mask13 = df['tourney_name'].str.contains('Monte Carlo')
    df.loc[mask13, 'tourney_name'] = 'Monte Carlo'
    
    mask14 = df['tourney_name'].str.contains('New York')
    df.loc[mask14, 'tourney_name'] = 'New York'
    
    mask15 = df['tourney_name'].str.contains('Rome')
    df.loc[mask15, 'tourney_name'] = 'Rome'
    
    mask16 = df['tourney_name'].str.contains('Shanghai')
    df.loc[mask16, 'tourney_name'] = 'Shanghai'
    
    mask17 = df['tourney_name'].str.contains('Toronto')
    df.loc[mask17, 'tourney_name'] = 'Toronto'
    
    mask18 = df['tourney_name'].str.contains('Janeiro')
    df.loc[mask18, 'tourney_name'] = 'Rio de Janeiro'
    
    return df



     



















In [117]:



     


df = replace_tourney_name(df)



     























Handle missing values¶






















For surface¶


















In [118]:



     


df[df['surface'].isna()]['tourney_name'].unique()



     






















    
    
Out[118]:








array(['Davis Cup', 'Perth', 'Bristol', 'Manchester', 'Cape Town',
       'Mexico City', 'Calcutta', 'Port Elizabeth', 'San Juan', 'Phoenix',
       'Cannes', 'Los Angeles SoCal Chps', 'Oakland', 'Woodside',
       'Helsinki', 'Saltsjoebaden', 'San Antonio Collegiate',
       'Montana Vermala', 'Senigallia', 'Quebec City', 'Bucharest',
       'Mamaia', 'Las Vegas', 'Madrid', 'Valencia', 'Cannes Chps',
       'Hanau', 'Tournament of Champions WCT', 'Roanoke', 'Dublin',
       'Cedar Grove', 'Christchurch', 'Oslo', 'New Orleans WCT', 'Omaha',
       'Paramus', 'Calgary', 'Salt Lake City', 'Washington', 'Tokyo WCT',
       'Tokyo', 'Freeport', 'Shreveport', 'New York', 'Istanbul',
       'Quebec WCT', 'Vancouver WCT', 'Casablanca WCT', 'Seattle',
       'Tanglewood', 'Montreal WCT', 'Alamo WCT', 'Los Angeles',
       'Des Moines', 'Kansas City', 'Sacramento', 'Quebec', 'New Delhi',
       'Djkarta'], dtype=object)
























In [119]:



     


# Repérer les valeurs vides dans la colonne 'surface'
missing_surface = df[df['surface'].isnull()]

# Parcourir les lignes avec une surface manquante
for index, row in missing_surface.iterrows():
    # Récupérer la valeur de 'tourney_name' pour cette ligne
    tourney_name = row['tourney_name']
    
    # Trouver les lignes avec le même 'tourney_name' qui ont une valeur de 'surface'
    matching_rows = df[(df['tourney_name'] == tourney_name) & ~(df['surface'].isnull())]
    
    # S'il y a des lignes correspondantes, prendre la première valeur de 'surface' et la remplacer dans la ligne d'origine
    if not matching_rows.empty:
        surface_value = matching_rows.iloc[0]['surface']
        df.at[index, 'surface'] = surface_value



     























For winner height¶


















In [120]:



     


# Repérer les valeurs vides dans la colonne 'winner_ht'
missing_winner_ht = df[df['winner_ht'].isnull()]

# Parcourir les lignes avec une 'winner_ht' manquante
for index, row in missing_winner_ht.iterrows():
    # Récupérer la valeur de 'winner_id' pour cette ligne
    winner_id = row['winner_id']
    
    # Rechercher la ligne correspondante dans le DataFrame 'players'
    player_info = players[players['player_id'] == winner_id]
    
    # Si la ligne correspondante existe dans le DataFrame 'players' et qu'elle a une valeur pour la colonne 'height'
    if not player_info.empty and not pd.isnull(player_info.iloc[0]['height']):
        # Récupérer la valeur de 'height' et remplacer la valeur NaN dans 'winner_ht'
        height_value = player_info.iloc[0]['height']
        df.at[index, 'winner_ht'] = height_value



     























For loser height¶


















In [121]:



     


# Repérer les valeurs vides dans la colonne 'winner_ht'
missing_loser_ht = df[df['loser_ht'].isnull()]

# Parcourir les lignes avec une 'winner_ht' manquante
for index, row in missing_loser_ht.iterrows():
    # Récupérer la valeur de 'winner_id' pour cette ligne
    loser_id = row['loser_id']
    
    # Rechercher la ligne correspondante dans le DataFrame 'players'
    player_info = players[players['player_id'] == loser_id]
    
    # Si la ligne correspondante existe dans le DataFrame 'players' et qu'elle a une valeur pour la colonne 'height'
    if not player_info.empty and not pd.isnull(player_info.iloc[0]['height']):
        # Récupérer la valeur de 'height' et remplacer la valeur NaN dans 'winner_ht'
        height_value = player_info.iloc[0]['height']
        df.at[index, 'loser_ht'] = height_value



     























For loser age¶


















In [122]:



     


# Convertir 'dob' en datetime si ce n'est pas déjà fait
players['dob'] = pd.to_datetime(players['dob'])

df['tourney_date'] = pd.to_datetime(df['tourney_date'])

# Calculer l'âge à partir de 'tourney_date' et 'dob'
def calculate_age(row):
    if pd.isnull(row['loser_age']):
        loser_id = row['loser_id']
        player_info = players[players['player_id'] == loser_id]
        if not player_info.empty and not pd.isnull(player_info.iloc[0]['dob']):
            dob = player_info.iloc[0]['dob']
            tourney_date = row['tourney_date']
            return (tourney_date - dob).days // 365
    return row['loser_age']

# Appliquer la fonction calculate_age pour remplir les valeurs manquantes de 'loser_age'
df['loser_age'] = df.apply(calculate_age, axis=1)



     























Drop remaining NaN¶


















In [123]:



     


df=df.dropna(subset=['surface','winner_ht','winner_age','loser_ht','loser_age'])



     























Second column filter to keep only the essentials¶


















In [125]:



     


df=df[['tourney_name','surface','winner_id','winner_hand','winner_ht','winner_age',
    'loser_id','loser_hand','loser_ht','loser_age','round']]



     























Creation of 2 dataframes : one for victories and one for losses¶






















By doing this, I duplicate my number of columns and put myself in a player versus opponent perspective. Each match will be interpreted from the winner's point of view as a victory and then from the loser's point of view as a defeat.



















In [147]:



     


df_winner = df.copy()
df_loser = df.copy()



     



















In [148]:



     


df.columns



     






















    
    
Out[148]:








Index(['tourney_name', 'surface', 'winner_id', 'winner_hand', 'winner_ht',
       'winner_age', 'loser_id', 'loser_hand', 'loser_ht', 'loser_age',
       'round'],
      dtype='object')




























Rename columns to have one player and his opponent¶


















In [149]:



     


df_winner=df_winner.rename(columns={'winner_id':'Player','winner_hand':'Player_Hand','winner_ht':'Player_Height',
                         'winner_age':'Player_Age','loser_id':'Opponent','loser_hand':'Opponent_Hand',
                          'loser_ht':'Opponent_Height','loser_age':'Opponent_Age'})



     























Players here are the winners so Result=1¶


















In [150]:



     


df_winner['Result']=1



     























Rename columns to have one player and his opponent¶


















In [151]:



     


df_loser=df_loser.rename(columns={'loser_id':'Player','loser_hand':'Player_Hand','loser_ht':'Player_Height',
                         'loser_age':'Player_Age','winner_id':'Opponent','winner_hand':'Opponent_Hand',
                          'winner_ht':'Opponent_Height','winner_age':'Opponent_Age'})



     























Players here are the losers so Result=0¶


















In [152]:



     


df_loser['Result']=0



     



















In [153]:



     


df_loser=df_loser[['tourney_name','surface','Player','Player_Hand','Player_Height','Player_Age','Opponent','Opponent_Hand','Opponent_Height','Opponent_Age','round','Result']]



     























Concat to get the final dataframe¶


















In [158]:



     


df=pd.concat([df_winner,df_loser])



     























Instead of leaving two columns for the same feature, create a differential.¶


















In [160]:



     


df



     






















    
    
Out[160]:











  
    

      
      tourney_name
      surface
      Player
      Player_Hand
      Player_Height
      Player_Age
      Opponent
      Opponent_Hand
      Opponent_Height
      Opponent_Age
      round
      Result
    

  
  
    

      0
      Brisbane
      Hard
      105453
      R
      178.0
      29.0
      106421
      R
      198.0
      22.8
      F
      1
    

    

      1
      Brisbane
      Hard
      106421
      R
      198.0
      22.8
      104542
      R
      188.0
      33.7
      SF
      1
    

    

      2
      Brisbane
      Hard
      105453
      R
      178.0
      29.0
      104871
      R
      188.0
      31.8
      SF
      1
    

    

      3
      Brisbane
      Hard
      104542
      R
      188.0
      33.7
      200282
      R
      183.0
      19.8
      QF
      1
    

    

      4
      Brisbane
      Hard
      106421
      R
      198.0
      22.8
      105683
      R
      196.0
      28.0
      QF
      1
    

    

      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
    

    

      191915
      Tour Finals
      Hard
      105453
      R
      178.0
      24.8
      104925
      R
      188.0
      27.4
      SF
      0
    

    

      191916
      Tour Finals
      Hard
      103819
      R
      185.0
      33.2
      104925
      R
      188.0
      27.4
      F
      0
    

    

      191917
      Davis Cup
      Clay
      104542
      R
      188.0
      29.5
      104527
      R
      183.0
      29.6
      RR
      0
    

    

      191918
      Davis Cup
      Clay
      103819
      R
      185.0
      33.2
      104792
      R
      193.0
      28.2
      RR
      0
    

    

      191919
      Davis Cup
      Clay
      104755
      R
      185.0
      28.4
      103819
      R
      185.0
      33.2
      RR
      0
    

  



311238 rows × 12 columns


























In [161]:



     


df['Hand_Opposition'] = df['Player_Hand'] + ' vs ' + df['Opponent_Hand']



     



















In [162]:



     


df['Height_Differential'] = df['Player_Height']-df['Opponent_Height']



     



















In [163]:



     


df['Age_Difference'] = df['Player_Age']-df['Opponent_Age']



     



















In [165]:



     


df=df[['tourney_name','surface','Player','Opponent','Hand_Opposition','Height_Differential','Age_Difference']]



     



















In [168]:



     


df.to_csv('df.csv')



     























Players reprocessing¶






















Sélection des joueurs présents dans top500¶






















Add Full Name¶


















In [166]:



     


players['Name'] = players['name_first'] + ' ' + players['name_last']



     























Keep only the players in the ATP Top 500 beginning of 2024¶


















In [167]:



     


players = players[players['Name'].isin(top500['Player'])]



     



















In [171]:



     


players=players.rename(columns={'Name':'Player'})



     























Get full data : Players characteristics and Rankings from ATP¶


















In [172]:



     


full_data=pd.merge(players,top500,on='Player',how='left')



     



















In [173]:



     


full_data



     






















    
    
Out[173]:











  
    

      
      player_id
      name_first
      name_last
      hand
      dob
      ioc
      height
      wikidata_id
      Player
      Rank
      Official Points
    

  
  
    

      0
      100644
      Alexander
      Zverev
      R
      1970-01-01 00:00:00.019970420
      GER
      198.0
      Q13990552
      Alexander Zverev
      6
      5085
    

    

      1
      104527
      Stan
      Wawrinka
      R
      1970-01-01 00:00:00.019850328
      SUI
      183.0
      Q193661
      Stan Wawrinka
      74
      797
    

    

      2
      104755
      Richard
      Gasquet
      R
      1970-01-01 00:00:00.019860618
      FRA
      185.0
      Q209436
      Richard Gasquet
      118
      538
    

    

      3
      104792
      Gael
      Monfils
      R
      1970-01-01 00:00:00.019860901
      FRA
      193.0
      Q186429
      Gael Monfils
      54
      937
    

    

      4
      104918
      Andy
      Murray
      R
      1970-01-01 00:00:00.019870515
      GBR
      190.0
      Q10125
      Andy Murray
      67
      845
    

    

      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
    

    

      185
      210097
      Ben
      Shelton
      L
      1970-01-01 00:00:00.020021009
      USA
      NaN
      Q108532383
      Ben Shelton
      17
      2055
    

    

      186
      210136
      Mark
      Lajal
      R
      1970-01-01 00:00:00.020030512
      EST
      NaN
      NaN
      Mark Lajal
      195
      308
    

    

      187
      210150
      Jakub
      Mensik
      U
      1970-01-01 00:00:00.020050901
      CZE
      NaN
      Q102228055
      Jakub Mensik
      87
      701
    

    

      188
      210234
      Juncheng
      Shang
      L
      1970-01-01 00:00:00.020050101
      CHN
      NaN
      Q106466705
      Juncheng Shang
      139
      465
    

    

      189
      210506
      Alex
      Michelsen
      U
      1970-01-01 00:00:00.020040825
      USA
      NaN
      NaN
      Alex Michelsen
      76
      776
    

  



190 rows × 11 columns


























In [174]:



     


full_data.to_csv('players_information.csv')



     



















In [403]:



     


to_predict = full_data[['player_id','Player','height','hand','dob','Rank','Official Points']]



     



















In [404]:



     


to_predict=to_predict.dropna(subset=['height'])



     



















In [405]:



     


to_predict['dob'] = to_predict['dob'].apply(convert_to_date)



     























Encodage des variables textuelles devant avoir les mêmes valeurs entre colonnes¶


















In [433]:



     


# Créer un ensemble de tous les ID uniques
unique_ids = set(df['Player_1_id']).union(set(df['Player_2_id'])).union(set(df['Winner']))

# Créer un dictionnaire pour mapper chaque ID unique à un nombre unique
id_to_number = {id_: i for i, id_ in enumerate(unique_ids)}

# Remplacer les valeurs des colonnes Player_1_id, Player_2_id et Winner par les nombres correspondants
df['Player_1_id'] = df['Player_1_id'].map(id_to_number)
df['Player_2_id'] = df['Player_2_id'].map(id_to_number)
df['Winner'] = df['Winner'].map(id_to_number)



     



















In [434]:



     


df



     






















    
    
Out[434]:











  
    

      
      tourney_name
      surface
      Player_1_id
      Player_2_id
      Hand_Opposition
      Height_Differential
      Age_Difference
      Ranking_Gap
      Ranking_Points_Difference
      Winner
    

  
  
    

      0
      Canada Masters
      Hard
      1724
      1926
      R vs R
      -21.0
      6.3
      -3.0
      -3.0
      29
    

    

      8
      Stuttgart
      Grass
      823
      818
      R vs R
      13.0
      -0.1
      -25.0
      -25.0
      29
    

    

      11
      Metz
      Hard
      205
      537
      R vs R
      -5.0
      4.9
      8.0
      8.0
      29
    

    

      13
      Chennai
      Hard
      907
      130
      L vs R
      -5.0
      -9.8
      -15.0
      -15.0
      378
    

    

      14
      Mexico City
      Clay
      159
      17
      R vs R
      0.0
      -2.4
      35.0
      35.0
      29
    

    

      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
    

    

      210737
      Stuttgart
      Clay
      221
      986
      R vs R
      10.0
      9.1
      -20.0
      -20.0
      29
    

    

      210738
      Washington
      Hard
      1995
      259
      R vs R
      15.0
      0.0
      -30.0
      -30.0
      29
    

    

      210739
      Toulouse
      Hard
      2023
      1812
      R vs R
      -5.0
      -2.1
      -67.0
      -67.0
      378
    

    

      210740
      Australian Open
      Hard
      626
      771
      R vs R
      5.0
      1.4
      -183.0
      -183.0
      29
    

    

      210741
      Metz
      Hard
      893
      912
      R vs R
      5.0
      0.2
      59.0
      59.0
      29
    

  



112951 rows × 10 columns


























In [435]:



     


numeric = ['Height_Differential','Age_Difference', 'Ranking_Gap', 'Ranking_Points_Difference']
string = ['tourney_name','surface','Hand_Opposition']



     



















In [436]:



     


for col in df.columns:
    if col in numeric:
        # Standardisation des données numériques
        scaler = StandardScaler()
        df[col] = scaler.fit_transform(df[[col]])
    elif col in string:
        # Encodage des données de chaîne
        encoder = LabelEncoder()
        df[col] = encoder.fit_transform(df[col])



     























Remise en forme de Winner¶


















In [452]:



     


df['Winner'] = df['Winner'].replace({29: 0, 378: 1})



     























Récupération de df¶


















In [453]:



     


df



     






















    
    
Out[453]:











  
    

      
      tourney_name
      surface
      Player_1_id
      Player_2_id
      Hand_Opposition
      Height_Differential
      Age_Difference
      Ranking_Gap
      Ranking_Points_Difference
      Winner
    

  
  
    

      0
      51
      3
      1724
      1926
      8
      -2.284599
      1.221184
      -0.048692
      -0.048692
      0
    

    

      8
      1857
      2
      823
      818
      8
      1.424725
      -0.023975
      -0.209432
      -0.209432
      0
    

    

      11
      1789
      3
      205
      537
      8
      -0.539035
      0.948806
      0.031678
      0.031678
      0
    

    

      13
      55
      3
      907
      130
      4
      -0.539035
      -1.911170
      -0.136369
      -0.136369
      1
    

    

      14
      1790
      1
      159
      17
      8
      0.006454
      -0.471454
      0.228950
      0.228950
      0
    

    

      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
    

    

      210737
      1857
      1
      221
      986
      8
      1.097432
      1.765942
      -0.172901
      -0.172901
      0
    

    

      210738
      1886
      3
      1995
      259
      8
      1.642921
      -0.004520
      -0.245964
      -0.245964
      0
    

    

      210739
      1876
      3
      2023
      1812
      8
      -0.539035
      -0.413088
      -0.516300
      -0.516300
      1
    

    

      210740
      17
      3
      626
      771
      8
      0.551943
      0.267859
      -1.363840
      -1.363840
      0
    

    

      210741
      1789
      3
      893
      912
      8
      0.551943
      0.034392
      0.404303
      0.404303
      0
    

  



112951 rows × 10 columns


























In [454]:



     


data = df.copy()



     



















In [455]:



     


data.to_csv('data.csv')



     



















In [456]:



     


df = data



     























Données d'entraînement et de test¶


















In [457]:



     


X=df.drop(columns=['Winner'])
y=df['Winner']



     



















In [458]:



     


X_train,X_test,y_train,y_test = train_test_split(X,y,test_size=0.2,random_state=123)



     























1er modèle : RandomForest¶


















In [459]:



     


rdf = RandomForestClassifier(n_estimators=100, random_state=42)



     



















In [460]:



     


rdf.fit(X_train,y_train)



     






















    
    
Out[460]:







RandomForestClassifier(random_state=42)
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. 
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.
RandomForestClassifier(random_state=42)
























In [461]:



     


y_pred = rdf.predict(X_test)



     



















In [462]:



     


print('Classification report:\n', classification_report(y_test, y_pred))



     






















    
    






Classification report:
               precision    recall  f1-score   support

           0       0.73      0.76      0.74     12509
           1       0.68      0.66      0.67     10082

    accuracy                           0.71     22591
   macro avg       0.71      0.71      0.71     22591
weighted avg       0.71      0.71      0.71     22591

























In [465]:



     


importances = rdf.feature_importances_
importances



     






















    
    
Out[465]:








array([0.13019673, 0.03287846, 0.14151982, 0.14300902, 0.02881643,
       0.09643659, 0.14001425, 0.14611234, 0.14101636])
























In [475]:



     


param_grid = {
    'n_estimators': [50, 100, 150],
    'max_depth': [None, 10, 20],
    'min_samples_split': [2, 5, 10],
    'min_samples_leaf': [1, 2, 4],
    'bootstrap': [True, False]
}



     



















In [476]:



     


# Instancier RandomizedSearchCV pour effectuer la recherche aléatoire sur la grille d'hyperparamètres
rdf_up = RandomizedSearchCV(estimator=rdf, param_distributions=param_grid, n_iter=50, cv=3, random_state=42, n_jobs=-1)



     



















In [477]:



     


rdf_up.fit(X_train,y_train)



     






















    
    
Out[477]:







RandomizedSearchCV(cv=3, estimator=RandomForestClassifier(random_state=42),
                   n_iter=50, n_jobs=-1,
                   param_distributions={'bootstrap': [True, False],
                                        'max_depth': [None, 10, 20],
                                        'min_samples_leaf': [1, 2, 4],
                                        'min_samples_split': [2, 5, 10],
                                        'n_estimators': [50, 100, 150]},
                   random_state=42)
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. 
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.
RandomizedSearchCV(cv=3, estimator=RandomForestClassifier(random_state=42),
                   n_iter=50, n_jobs=-1,
                   param_distributions={'bootstrap': [True, False],
                                        'max_depth': [None, 10, 20],
                                        'min_samples_leaf': [1, 2, 4],
                                        'min_samples_split': [2, 5, 10],
                                        'n_estimators': [50, 100, 150]},
                   random_state=42)
RandomForestClassifier(random_state=42)
RandomForestClassifier(random_state=42)
























In [478]:



     


y_pred = rdf_up.predict(X_test)



     



















In [479]:



     


accuracy = accuracy_score(y_test, y_pred)
precision = precision_score(y_test, y_pred)
recall = recall_score(y_test, y_pred)

# Afficher les performances
print("Accuracy: {:.2f}".format(accuracy))
print("Precision: {:.2f}".format(precision))
print("Recall: {:.2f}".format(recall))



     






















    
    






Accuracy: 0.71
Precision: 0.67
Recall: 0.67




























2e modèle : xgBoost¶


















In [466]:



     


# Créer un objet DMatrix pour l'entraînement et le test
dtrain = xgb.DMatrix(X_train, label=y_train)
dtest = xgb.DMatrix(X_test, label=y_test)



     



















In [467]:



     


# Définir les paramètres du modèle
params = {
    'objective': 'binary:logistic',  # pour un problème de classification binaire
    'eval_metric': ['logloss', 'error'],  # métriques d'évaluation
    'max_depth': 3,  # profondeur maximale de l'arbre
    'eta': 0.1,  # taux d'apprentissage
    'gamma': 0.5  # paramètre de réduction de perte minimale
}



     



















In [468]:



     


# Entraîner le modèle
num_rounds = 100  # nombre d'itérations
model = xgb.train(params, dtrain, num_rounds)



     



















In [469]:



     


# Faire des prédictions sur l'ensemble de test
y_pred = model.predict(dtest)
y_pred_binary = [1 if p >= 0.5 else 0 for p in y_pred]  # conversion en classes binaires



     



















In [470]:



     


accuracy = accuracy_score(y_test, y_pred_binary)
precision = precision_score(y_test, y_pred_binary)
recall = recall_score(y_test, y_pred_binary)

# Afficher les performances
print("Accuracy: {:.2f}".format(accuracy))
print("Precision: {:.2f}".format(precision))
print("Recall: {:.2f}".format(recall))



     






















    
    






Accuracy: 0.66
Precision: 0.62
Recall: 0.60
























In [ ]:



     


 



     



















In [ ]:



     


 



     



















In [ ]:



     


 



     



















In [ ]:



     


 



     



















In [471]:



     


to_predict



     






















    
    
Out[471]:











  
    

      
      player_id
      Player
      height
      hand
      dob
      Rank
      Official Points
    

  
  
    

      0
      100644
      Alexander Zverev
      198.0
      R
      1997-04-20
      6
      5085
    

    

      1
      104527
      Stan Wawrinka
      183.0
      R
      1985-03-28
      74
      797
    

    

      2
      104755
      Richard Gasquet
      185.0
      R
      1986-06-18
      118
      538
    

    

      3
      104792
      Gael Monfils
      193.0
      R
      1986-09-01
      54
      937
    

    

      4
      104918
      Andy Murray
      190.0
      R
      1987-05-15
      67
      845
    

    

      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
    

    

      168
      208502
      Dominic Stricker
      180.0
      L
      2002-08-16
      112
      557
    

    

      171
      209070
      Arthur Cazaux
      183.0
      R
      2002-08-23
      86
      701
    

    

      172
      209098
      Hamad Medjedovic
      185.0
      R
      2003-07-18
      116
      541
    

    

      180
      209857
      Leandro Riedi
      191.0
      R
      2002-01-27
      160
      396
    

    

      181
      209950
      Arthur Fils
      185.0
      R
      2004-06-12
      44
      1028
    

  



146 rows × 7 columns


























In [63]:



     


df



     






















    
    
Out[63]:











  
    

      
      tourney_id
      surface
      tourney_level
      winner_id
      winner_hand
      winner_age
      loser_id
      loser_hand
      loser_ht
      loser_age
      winner_rank
      winner_rank_points
      loser_rank
      loser_rank_points
      Result
    

  
  
    

      0
      7825
      3
      0
      105453
      2
      0.821557
      106421
      2
      2.062667
      -0.697878
      -0.547538
      1.279864
      -0.595793
      1.117571
      1
    

    

      1
      7825
      3
      0
      106421
      2
      -0.708139
      104542
      2
      0.560875
      1.904041
      -0.489616
      0.347367
      0.775595
      -0.670376
      1
    

    

      2
      7825
      3
      0
      105453
      2
      0.821557
      104871
      2
      0.560875
      1.450495
      -0.547538
      1.279864
      -0.448199
      0.184860
      1
    

    

      3
      7825
      3
      0
      104542
      2
      1.981165
      200282
      2
      -0.190021
      -1.414003
      1.355633
      -0.679940
      -0.503547
      0.434388
      0
    

    

      4
      7825
      3
      0
      106421
      2
      -0.708139
      105683
      2
      1.762309
      0.543404
      -0.489616
      0.347367
      -0.583493
      0.994820
      1
    

    

      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
    

    

      191915
      7056
      3
      2
      104925
      2
      0.426797
      105453
      2
      -0.940917
      -0.220462
      -0.613736
      4.991350
      -0.663440
      3.781885
      1
    

    

      191916
      7056
      3
      2
      104925
      2
      0.426797
      103819
      2
      0.110337
      1.784686
      -0.613736
      4.991350
      -0.681889
      7.881990
      1
    

    

      191917
      7084
      1
      1
      104527
      2
      0.969592
      104542
      2
      0.560875
      0.901466
      -0.588912
      2.265547
      -0.620392
      1.885272
      1
    

    

      191918
      7084
      1
      1
      104792
      2
      0.624177
      103819
      2
      0.110337
      1.784686
      -0.464792
      0.259494
      -0.681889
      8.888151
      0
    

    

      191919
      7084
      1
      1
      103819
      2
      1.857802
      104755
      2
      0.110337
      0.638887
      -0.605461
      4.812135
      -0.534296
      0.486709
      1
    

  



102934 rows × 15 columns


























In [ ]: